Subject: Re: Edlabel help, MV2K
To: None <,>
From: Carl Lowenstein <>
List: port-vax
Date: 06/29/1998 09:39:05
> Date: Sat, 27 Jun 1998 08:16:11 -0400
> From: "Sheila H. //or//Bob (depends on who is writing" <>
> MIME-Version: 1.0
> Subject: Re: Edlabel help, MV2K
> Hi Roger,
> I don't know if my experience will help or not - my set up is a uVII
> (ka630) with a CMD scsi controller.
> I have not been able to get a fujitus drive to work - a 600 mB drive - I
> have installed Sun drives (rmarked hp330MB), and HP 1.3GB drives on the
> same controller, with minimal problem AFTER I GOT MY TERMINATION STUFF
> I also have a USD scsi controller.  I was not able to make the fuji work
> on that one either.
> Take the fuji and plug it into a Sun, no problem.  Maybe this will jog
> some folks' memories.
> Roger Ivie wrote:
> > 
> > OK, so I've given up on the 3.5" diskette and attached an honest-to-god
> > RX33 drive to the thing. This lets me bring up the bootloader, at which
> > point I can run rom()edlabel to bring up edlabel. I have a big ol' Fujitsu
> > 600MB disk (chosen because I know it works) connected to the machine,
> > jumpered as drive 2. Playing with edlabel, I find it seems to want to
> > refer to this drive as sd(2,0,0) rather than the documented
> > device(ctrl,adapter,disk,part). When I give edlabel sd(2,0,0), I see
> > a bunch of "phase mismatch" messages and "SCSI: busy error" messages.
> > During all this, the light on the drive blinks, so I know it's selecting the
> > drive. A log is attached. Does it ring any bells?

Possible bell ringing:  this seems very similar to problems that us
NeXT users have had with this vintage of Fujitsu disk drives.  The
SCSI controller can't cope with disk-initiated synchronous transfer
negotiation.  I think the Fuji drive has a jumper to disable this.
And Sun SCSI controllers know about synchronous transfer.


        carl lowenstein         marine physical lab     u.c. san diego
        {decvax|ucbvax} !ucsd!mpl!cdl